Secondary VM in a fault tolerance configuration reports using a higher percentage of memory than primary

[tdupont]

I recently created a fault tolerant domain controller for our VM environment and I noticed that the secondary VM is reporting that it is using 87% of configured memory even though it is using the same MB of memory as the primary.

I have the VM configured to use 2GB of memory. The Primary is reporting that it is using 755MB and the secondary is reporting that it is using 768MB of memory but when you look at the column that displays guest memory percentage the primary is reporting 5% usage while the secondard is reporting 87% usage.

Is this normal behavior? I would think that they would be configured and reporting nearly identical conditions.

[tdupont]

After running the VM for a week, there really are no changes in the memory reporting...

Primary: 904MB Host Mem and 6% Guest memory usage

Secondary : 908MB Host Mem and 66% Guest memory usage

I am occasionally getting alerts about memory usage on the secondary exceeding 85% but the Primary VM is not going above 50%

[paul_xtravirt]

Yes, in my experience that is correct, I believe there is additional overhead due to the memory copys from the first to the second taking place (Ithink)

It should say in the Fault Tolerance documents somewhere

If you found this helpful, please consider awarding some points

[tdupont]

I guess my confusion comes from large difference in Guest Memory % being reported.

It almost seems to me that the secondary VM is configured to use less memory than the primary, driving up the Guest Memory % being used while keeping the same amount of Host memory being used.

I would expect the secondary guest VM to use more memory, but I did not expect it to be using 11x more.

I may just be misreading the intent of these metrics as well.

Thank you for the information.

[tdupont]

What is really annoying about this issue, is that we have a global alarm for VM memory usage, and this is the only VM that triggers it on a regular basis (when the memory usage exceeds 85% for longer than 10 minutes to go to yellow and more than 95% for 10 minutes to go red) It goes red at least a couple times a day. The Primary VM is not even a busy server. It is a domain controller and nothing else. Our entire environment only has 800 users and about the same number of servers/computers.

To add insult to injury, it looks like I will have to reconfigure all of my alarms just to make an exception for this single VM.

I guess I am just stubborn, but why should the secondary be using so MUCH more memory than the Primary, when the system is basically at idle?

[MascoIT]

Hi,

Did you ever find a satisfactory resolution to this? I recently started tinkering with FT and am seeing the same issue. VMware suport gave me a 'we think that's normal behavior'. It didn't sound lke anybody tried it in the lab.